Ann Nucl Med (2014) 28:455–462 DOI 10.1007/s12149-014-0836-7
ORIGINAL ARTICLE
Characteristics and trends of research on positron emission tomography: a bibliometric analysis, 2002–2012 Sora Baek • Dae Young Yoon • Kyung Joon Min Kyoung Ja Lim • Young Lan Seo • Eun Joo Yun
•
Received: 24 January 2014 / Accepted: 27 February 2014 / Published online: 12 March 2014 Ó The Japanese Society of Nuclear Medicine 2014
Abstract Objective We performed a bibliometric analysis of scientific publications focused on positron emission tomography (PET) over a 10-year period. Methods The MEDLINE and ISI Web of Knowledge databases were searched for English language original articles focused on PET in SCI/SCIE-indexed journals in 2002, 2007, and 2012. We selected the documents with titles that included ‘‘PET’’ or ‘‘positron emission.’’ The following information was obtained from each article: journal (year of publication, title, subject category, and impact factor), subspecialty, imaging modality used, tracer, species, sample size, number of authors, affiliation of the first author, declared funding, and country of origin. Results The yearly publication on PET increased from 547 (2002) to 986 (2007), and 1838 (2012). A total of 1753 (52.0 %) articles were published in journals in the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ category, 1512 (44.9 %) were in the subspecialty of oncology, 3245 (96.3 %) used PET or PET/CT, 1698 (50.4 %) used 2-fluoro-2-deoxy-D-glucose (FDG) as the radiotracer, 2378 (70.5 %) were human studies, 1294 (38.4 %) had a sample size of \20, 1674 (49.7 %) had [7 authors, 779 (23.1 %) were written by a first author from a department of nuclear medicine, and 1337 (39.7 %) were supported by
S. Baek Department of Nuclear Medicine, Kangdong Seong-Sim Hospital, Hallym University College of Medicine, 445 Gil-dong Kangdong-Gu, Seoul 134-701, Korea D. Y. Yoon (&) K. J. Min K. J. Lim Y. L. Seo E. J. Yun Department of Radiology, Kangdong Seong-Sim Hospital, Hallym University College of Medicine, 445 Gil-dong Kangdong-Gu, Seoul 134-701, Korea e-mail: [email protected]
government funding. The United States published 948 studies (28.1 %) followed by Japan (345, 10.2 %) and Germany (335, 9.9 %). In the time trend analysis oncology subspecialty, PET/MR as the imaging modality, FDG as the tracer, sample size [50, number of authors [7, radiology department affiliation of the first author, and government funding exhibited significantly positive trends. Conclusions The number of publication concerning PET has increased rapidly over the last decade. This bibliometric analysis revealed characteristics and trends of current PET research that provides useful information to researchers. Keywords
PET Research Publications Bibliometrics
Introduction Positron emission tomography (PET) is a diagnostic modality in nuclear medicine that enables a three-dimensional imaging of functional changes in the body [1]. Recently, PET has emerged as one of the most promising imaging modalities in clinical medicine. PET has been used in combination with high-resolution anatomical imaging modalities (i.e., CT or MR) to provide functional and anatomical information simultaneously [2, 3]. Many investigators, particularly in the last decade, have published articles on PET, reflecting the increased clinical use, technological progress, and expanded clinical applications of this technique. Bibliometry is a research method used to evaluate the quantity and quality of scientific research production within a specific field or a specific topic over a given period [4]. Bibliometric analysis provides insight into research patterns and trends in scientific publications. Although
123
456
several bibliometric studies have been conducted in the field of nuclear medicine [5–7], to the best of our knowledge, no study has explored the characteristics of research in PET. The purpose of this study was as follows: (1) to quantify the worldwide output, (2) to assess the characteristics, and (3) to determine the trends of PET research over a 10-year period (2002–2012).
Methods Our study was a retrospective bibliometric analysis that did not involve human subjects and was exempt from Institutional Review Board approval. Study search and data collection The first step consisted of identifying all the articles focused on PET during three 1-year periods (2002, 2007, and 2012). The National Library of Medicine’s PubMed database (URL: http://www.ncbi.nlm.nih.gov/pubmed/) was searched between February and April 2013 to identify the primary studies of interest. We selected documents including the words ‘‘PET’’ or ‘‘positron emission’’ in the title. The term ‘‘positron emission’’ was selected to widen our search to include articles with a variety of words in the title (‘‘positron emission tomography’’, ‘‘positron emission tomographic’’, ‘‘positron emission mammography’’). To establish the characteristics of the articles focused on PET, only original articles were included in the analysis. Original articles were considered to be documents that investigated clearly stated objectives or hypotheses and contained specifically articulated methods and results sections. Other document types such as case reports, review articles, commentaries, editorials, letters, technical notes, quizzes, lectures, book reviews, and news were excluded from the analysis. We included only articles written in English; articles published in other languages were excluded. Finally, we excluded articles with a focus on pet (a domesticated animal kept for companionship or amusement). Abstracting data on journals We accessed the Institute for Scientific Information (ISI) Web of Knowledge (URL: http://www.isiwebofknowledge. com), a database of more than 8400 journals in more than 170 scientific disciplines, for a list of all the journals indexed by the Science Citation Index (SCI) and the Science Citation Index Expanded (SCIE). The articles published in journals that were not indexed by SCI/SCIE were excluded from the analysis.
123
Ann Nucl Med (2014) 28:455–462
The analyzed journals (SCI/SCIE-indexed journals that published at least one original article focused on PET) were assigned to one of a number of subject categories (169, 171, and 176 categories in 2002, 2007, and 2012, respectively) based on information in the ISI Web of Knowledge-Journal Citation Reports (JCR). According to the ISI Web of Knowledge-JCR, a number of journals were listed under more than one subject category, depending on the content. Journals that were in more than one subject category were manually reviewed and assigned to one category based on the title of the journal, information contained in ‘‘Instructions to Authors,’’ and the table of contents of a sample issue. Decisions about assigning these journals were reached by consensus between two study investigators (S.B., a board-certified nuclear medicine physician with 3 years of experience and D.Y.Y., a board-certified radiologist with 18 years of experience). The journal impact factor (IF) of each journal was obtained for the year of publication (2002, 2007, and 2012) from the ISI Web of Knowledge-JCR. The IF of a journal for a given year is calculated by dividing the number of citations in the JCR year by the total number of articles published in the past two years [8]. Abstracting data on articles The following information was abstracted from each article for analysis: (1) the subspecialty [oncology, cardiovascular disease, inflammation/infection, metabolism, neurology, physics/instrumental, radiopharmaceutical (tracer), molecular imaging, or others (not conforming to one of the above categories)]; (2) the imaging modality used (PET or PET/ CT, PET/MR, or others); (3) the tracer [FDG (2-fluoro-2deoxy-D-glucose), others (including multiple tracers), or none]; (4) the species (human, animal, or others); (5) the sample size (none, \20, 21–50, or [50); (6) the number of authors (\4, 4–7, or [7); (7) the affiliation of the first author [radiology (including radiology, imaging center, and other imaging-related specialties), nuclear medicine, medicine or related specialties, surgery or related specialties, or other]; (8) the declared funding (government funding, other resources, or none); and (9) the country of origin. For the articles arising from a collaboration involving more than one country, the country of the first author was considered the country of origin of the paper. If the first author had an affiliation with more than one country, the country of origin of the paper was determined by the country of the corresponding author. Two investigators (S.B. and D.Y.Y.) initially reviewed the same 200 articles independently to ensure consistency of the data abstraction. Any disagreements were resolved in a consensus meeting; however, only two (1.0 %)
Ann Nucl Med (2014) 28:455–462
disagreements occurred with regard to the subspecialty field of research. After initial pilot abstraction, the total number of articles was divided randomly into 2 samples and manually reviewed by the investigators. The questionable cases were decided by consensus of the 2 study investigators.
457 Table 1 Comparison of SCI/SCIE-indexed journals which published original articles focused on PET in 2002, 2007, and 2012 2002
2007
2012
No. of articles
547
986
1838
No. of journals
178
247
378 3.8 ± 2.6*
Journal IF
Statistical analysis The data were recorded by publication year (2002, 2007, and 2012) and downloaded into a spreadsheet for descriptive statistics using Microsoft Excel version 11.0 (Microsoft Corporation, Redmond, WA, USA). The journal IFs were compared using Student’s t test and the other categorical variables were compared using the Chi-square test. The statistical analyses were performed using SPSS software (SPSS version 19.0 for Windows; SPSS, Chicago, IL, USA) and a p value of \0.05 was considered to be statistically significant.
Results A total of 547, 986, and 1838 original articles focused on PET were written in English and published in SCI/SCIEindexed journals in 2002, 2007, and 2012, respectively. The absolute number of articles increased 3.4-fold from 2002 through 2012 (Table 1). The articles were published in 515 scientific journals. Table 2 presents a list of the 20 journals with the highest number of articles focused on PET. Regarding the JCR subject categories of the journals, 1753 articles (52.0 %) were published in journals under the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ subject category. Other subject categories included ‘‘Oncology’’ (n = 269, 8.0 %) and ‘‘Neurosciences’’ (n = 254, 7.5 %) (Table 3). The mean IF of the journals that published PET research increased from 3.3 in 2002 to 3.8 in 2007, a statistically significant increase (p \ 0.05). No significant change in the IF was observed from 2007 to 2012 (Table 1). ‘Oncology’ was found to be the most productive field (n = 1512, 44.9 %), followed by ‘neurology’ (n = 625, 18.5 %), ‘physics/instrumental’ (n = 325, 9.6 %), and ‘radiopharmaceutical’ (n = 237, 7.0 %). In the time trend analysis, the ‘oncology’, ‘radiopharmaceutical’ and ‘molecular imaging’ subspecialties demonstrated a significant increase, whereas the ‘neurology’ subspecialty demonstrated a significant decrease (Table 4). Other variables including the imaging modality used, tracer, species, sample size, number of authors, affiliation of the first author, and declared funding are summarized in Table 4. The variables that showed a statistically significant increase over time included the following: PET/MR in
Mean ± SD
3.3 ± 2.2
3.8 ± 2.4*
25 percentile Journal IF
1.924
2.209
2.247
50 percentile (median) journal IF
2.930
3.561
3.095
75 percentile journal IF
4.587
5.457
5.114
IF impact factor * Statistically significant difference compared to 2002 (p = 0.001, Student’s t test)
imaging modality, FDG in tracer, sample size [ 50, number of authors [ 7, department of radiology in the affiliation of the first author, and government funding. The variables that showed a statistically significant decrease over time included the following: others in tracer, sample size \ 20, number of authors \ 4 and 4–7, nuclear medicine in the affiliation of the first author, and other resources of declared funding (Table 4). Table 5 shows the data from the 20 most productive countries in PET research. The United States (US) was dominant with the highest number of publications (n = 948, 28.1 %) followed by Japan (n = 345, 10.2 %), Germany (n = 335, 9.9 %), South Korea (n = 178, 5.3 %), and the United Kingdom (UK) (n = 174, 5.2 %). The articles from these five countries accounted for 58.7 % of the PET research. South Korea, Italy, China, Taiwan, Turkey, and India showed a statistically significant increase over the analyzed period in the proportion of articles, whereas US, Germany, the UK, and Finland exhibited a significant decline in the percentage of articles (Table 5).
Discussion PET is an imaging technique that provides detailed information regarding the molecular and metabolic changes of an organ. In the mid-1980s, with the development of commercial PET scanners, PET moved from the research field to clinical practice. A major advance in PET scanning occurred in the early 1990s with the introduction of the PET/CT scanner. The hybrid of PET and CT imaging, simultaneously showing metabolic and structural images, enhances the ability of a physician to localize abnormalities accurately. The addition of integrated dual-modality imaging and the development of new radiotracers have accelerated the rapid growth of PET during recent years [9]. It is estimated that in the US, the number of PET/CT
123
458 Table 2 Top 10 SCI/SCIEindexed journals with the highest number of articles focused on PET in 2002, 2007, and 2012
Table 3 Top 10 subject categories with the highest number of articles focused on PET in 2002, 2007, and 2012
Values in parentheses are percentages Ranking based on the total number of original articles published in 2002, 2007, and 2012 :
Statistically significant increase between 2002 and 2012 (Chi-square test)
;
Statistically significant decrease between 2002 and 2012 (Chi-square test)
Ann Nucl Med (2014) 28:455–462
Rank
2002
2007
Journal
No. of articles
Journal
No. of articles
Journal
No. of articles
1
Eur J Nucl Med Mol Imaging
65
J Nucl Med
110
J Nucl Med
153
2
J Nucl Med
49
Eur J Nucl Med Mol Imaging
106
Eur J Nucl Med Mol Imaging
135
3
Nucl Med Biol
28
Phys Med Biol
30
4
Neuroimage
25
Neuroimage
29
Ann Nucl Med
55
5
Phys Med Biol
17
Nucl Med Commun
28
Clin Nucl Med
52
6
Synapse
16
Int J Radiat Oncol Biol Phys
27
Phys Med Biol
51
7
Acta Radiol
14
Nucl Med Biol
26
Eur J Radiol
49
8
J Cereb Blood Flow Metab
13
Mol Imaging Biol
22
Nucl Med Biol
49
9
13
Ann Nucl Med
21
9
J Cereb Blood Flow Metab
18
Int J Radiat Oncol Biol Phys Mol Imaging Biol
47
10
Nucl Med Commun Ann Nucl Med
Rank
Category
Nucl Med Commun
56
42
No. of Articles 2002
2007
2012
Total
239 (43.7)
518 (52.5)
996 (54.2)
1753 (52.0)
1
Radiology, Nuclear Medicine & Medical Imaging:
2
Oncology
45 (8.2)
90 (9.1)
134 (7.3)
269 (8.0)
3
Neurosciences;
77 (14.1)
87 (8.8)
90 (4.9)
254 (7.5)
4
Neuroimaging;
27 (4.9)
33 (3.3)
48 (2.6)
108 (3.2)
5
Clinical Neurology;
27 (4.9)
29 (2.9)
42 (2.3)
98 (2.9)
5
Chemistry, Inorganic & Nuclear
8 (1.5)
25 (2.5)
52 (2.9)
85 (2.5)
6 7
Pharmacology & Pharmacy Surgery
12 (2.2) 14 (2.6)
20 (2.0) 12 (1.2)
42 (2.3) 36 (2.0)
74 (2.2) 62 (1.8)
8
Hematology
3 (0.5)
19 (1.9)
31 (1.7)
53 (1.6)
9
Otorhinolaryngology
5 (0.9)
13 (1.3)
33 (1.8)
51 (1.5)
10
Cardiac & Cardiovascular Systems
10 (1.8)
19 (1.9)
21 (1.1)
50 (1.5)
systems increased dramatically by more than 10-fold (from approximately 200 to 2085) and the number of PET examinations increased nearly sevenfold (from approximately 250000 to approximately 1.74 million) between 2001 and 2010 [10]. In the evidence-based model of clinical practice, research precedes clinical adoption. One of the striking findings of our study was the rapid growth in the number of publications on PET research during the last decade. Overall, the total number of publications increased 3.4-fold from 2002 through 2012. This marked growth in the scientific publications in this field suggests that PET is an expanding and promising field of research. The increase in
123
2012
PET publications appears to be more pronounced than those observed in other imaging modalities (i.e., ultrasonography or MR) [11, 12]. This growth may reflect the following: (1) a rapidly increasing worldwide commercial manufacture of PET/CT systems, (2) a widening application of PET resulting from technological development and clinical introduction of new pharmacological agents, and (3) an increasing knowledge and scientific interest in PET. Our study showed that there are a variety of journals that publish PET research. The articles on PET were predominantly published in imaging journals (journals in the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ category). An analysis of the journals with the highest number
Ann Nucl Med (2014) 28:455–462
459
Table 4 Characteristics and trends of original articles focused on PET in 2002, 2007, and 2012
Number of Articles 2002
2007
2012
Total
Subspecialty Oncology:
191 (34.9)
436 (44.2)
885 (48.2)
1512 (44.9)
Cardiovascular disease
28 (5.1)
56 (5.7)
129 (7.0)
213 (6.3)
Inflammation/infection
13 (2.4)
35 (3.5)
58 (3.2)
106 (3.1)
Metabolism
19 (3.5)
37 (3.8)
42 (2.3)
98 (2.9)
Neurology;
193 (35.3)
216 (21.9)
216 (11.8)
625 (18.5)
Physics/Instrumental
52 (9.5)
96 (9.7)
177 (9.6)
325 (9.6)
Radiopharmaceutical (tracer):
32 (5.9)
41 (4.2)
164 (8.9)
237 (7.0)
Molecular imaging:
13 (2.4)
59 (6.0)
77 (4.2)
149 (4.4)
6 (1.1)
10 (1.0)
90 (4.9)
106 (3.1)
534 (97.6)
951 (96.5)
1760 (95.8)
3245 (96.3)
2 (0.4)
10 (1.0)
52 (2.8)
64 (1.9)
11 (2.0)
25 (2.5)
26 (1.4)
62 (1.8)
236 (43.1)
493 (50.0)
969 (52.7)
1698 (50.4)
278 (50.8)
432 (43.8)
761 (41.4)
1471 (43.6)
33 (6.0)
61 (6.2)
108 (5.9)
202 (6.0)
Human
390 (71.3)
696 (70.6)
1292 (70.3)
2378 (70.5)
Animal
80 (14.6)
159 (16.1)
255 (13.9)
494 (14.7)
Others
77 (14.1)
131 (13.3)
291 (15.8)
496 (14.7)
Others: Imaging modality used PET or PET/CT PET/MR: Others Tracer FDG*, Others
: ;
None Species
Sample size None
64 (11.7)
132 (13.4)
232 (12.6)
428 (12.7)
\20;
296 (54.1)
391 (39.7)
607 (33.0)
1294 (38.4)
21–50
131 (23.9)
244 (24.7)
456 (24.8)
831 (24.7)
:
56 (10.2)
219 (22.2)
543 (29.5)
818 (24.3)
\4;
74 (13.5)
66 (6.7)
118 (6.4)
258 (7.7)
4–7;
260 (47.5)
441 (44.7)
738 (40.2)
1439 (42.7)
[7:
213 (38.9)
479 (48.6)
982 (53.4)
1674 (49.7)
[50 Number of authors
Affiliation of the first author Radiology (imaging center):
Values in parentheses are percentages * 2-fluoro-2-deoxy-D-glucose :
Statistically significant increase between 2002 and 2012 (Chi-square test)
;
Statistically significant decrease between 2002 and 2012 (Chi-square test)
89 (16.3)
242 (24.5)
393 (21.4)
724 (21.5)
Nuclear medicine;
155 (28.3)
255 (25.9)
369 (20.1)
779 (23.1)
Medicine or related specialties
126 (23.0)
223 (22.6)
374 (20.3)
723 (21.4)
56 (10.2)
82 (8.3)
173 (9.4)
311 (9.2)
121 (22.1)
184 (18.7)
529 (28.8)
834 (24.7)
Government funding:
196 (35.8)
379 (38.4)
762 (41.5)
1337 (39.7)
Other resources;
109 (19.9)
158 (16.0)
240 (13.1)
507 (15.0)
None
242 (44.2)
449 (45.5)
836 (45.5)
1527 (45.3)
Surgery or related specialties Others
:
Declared funding
Total
of articles in this field showed a prevalence of articles in journals focused on ‘‘nuclear medicine’’. Five of the 10 journals with the highest number of articles contained the word ‘‘nuclear’’ in the journal title in 2002, 2007, and 2012 (Table 2). Approximately one-half (48 %) of the articles
547
986
1838
3371
on PET were published in non-imaging journals (journals under the category of other than ‘‘Radiology, Nuclear Medicine & Medical Imaging’’). This tendency primarily reflects the impact of PET on basic and clinical specialties. Fifty-five percent of the articles on PET were primarily
123
460 Table 5 Top 20 countries ranked according to total number of original articles focused on PET in 2002, 2007, and 2012
Ann Nucl Med (2014) 28:455–462
Rank
Number of Articles 2002
2007
2012
Total 948 (28.1)
1
United States;
163 (29.8)
319 (32.4)
466 (25.4)
2
Japan
62 (11.3)
98 (9.9)
185 (10.1)
345 (10.2)
3
Germany;
61 (11.2)
120 (12.2)
154 (8.4)
335 (9.9)
4
South Korea:
11 (2.0)
42 (4.3)
125 (6.8)
178 (5.3)
a, ;
5
United Kingdom
45 (8.2)
35 (3.5)
94 (5.1)
174 (5.2)
6
France
18 (3.3)
52 (5.3)
81 (4.4)
151 (4.5)
Values in parentheses are percentages
7
Italy:
10 (1.8)
30 (3.0)
97 (5.3)
137 (4.1)
8
The Netherlands
23 (4.2)
38 (3.9)
73 (4.0)
134 (4.0)
Ranking based on the total number of original articles published in 2002, 2007, and 2012
9
Canada
18 (3.3)
34 (3.4)
55 (3.0)
107 (3.2)
10
Chinab,
1 (0.2)
20 (2.0)
83 (4.5)
104 (3.1)
11 12
Switzerland Sweden
22 (4.0) 21 (3.8)
28 (2.8) 22 (2.2)
44 (2.4) 43 (2.3)
94 (2.8) 86 (2.6)
13
Denmark
12 (2.2)
20 (2.0)
42 (2.3)
74 (2.2)
14
Belgium
13 (2.4)
18 (1.8)
31 (1.7)
62 (1.8)
15
Australia
12 (2.2)
10 (1.0)
33 (1.8)
55 (1.6)
5 (0.9)
10 (1.0)
38 (2.1)
53 (1.6)
3 (0.5)
11 (1.1)
31 (1.7)
45 (1.3)
15 (1.5)
13 (0.7)
41 (1.2)
a
Includes articles originating from England, Scotland, Wales, and Northern Ireland
b
Includes articles originating from Hong Kong
: Statistically significant increase between 2002 and 2012 (Chi-square test) ; Statistically significant decrease between 2002 and 2012 (Chi-square test)
:
:
16
Taiwan
17
Spain
18
Finland;
13 (2.4)
19
Turkey:
1 (0.2)
9 (0.9)
27 (1.5)
37 (1.1)
20
India:
0 (0)
1 (0.1)
34 (1.8)
35 (1.0)
authored by non-radiologist/non-nuclear medicine physicians. Publication across disciplines is a current trend in scientific literature [13–16]. The most commonly used indicator to evaluate the relative quality of scientific journals is the IF of a journal, but it is not, however, a perfect tool to measure the quality of journals [17, 18]. The IF of biomedical journals has generally shown a tendency to rise, according to the expansion in the size of the relevant literature [19, 20]. In our study, the IF of journals that published PET research did not change significantly from 2007 to 2012. Considering the overall increase in the IF of the journal, this finding indicates that the scientific impact of PET research decreased between 2007 and 2012. The proportion of PET articles published in the nuclear medicine journal with the highest impact factor (Journal of Nuclear Medicine) decreased from 11.2 % in 2007 to 8.3 % in 2012. The analysis of the distribution of papers according to the subspecialty field revealed that oncology accounted for almost one-half of PET research. This finding is predominantly based on the emerging role of PET or PET/CT using FDG in the detection of primary and recurrent malignant tumors. A total of 45 % of the articles on PET were primarily authored by radiologists or nuclear medicine physicians between 2002 and 2012. The percentage of articles written by radiologists (or physicians of the imaging center)
123
significantly increased, whereas the percentage of articles written by nuclear medicine physicians significantly decreased. The reason for this disparity is unclear and likely multifactorial. First, department of nuclear medicine has been partially or fully separated from radiology departments in most countries. In recent years, the borders between the two specialties have not been clearly defined (e.g., the emergence of imaging centers). Second, the use of hybrid machines such as PET/CT or PET/MR has resulted in a merger of nuclear medicine and radiology in PET research. The overall funding rate of 54.7 % in PET research over the entire study period is substantially higher than that reported for general radiology research (26.9 %) [21]. The positive association between funding and research productivity is well established [22]. We found that government funding for PET research increased nearly 6 % during the study period whereas the overall proportion of funding did not change significantly. Increased government funding such as National Institutes of Health may have led to an increase in financial support for higher-level research. The US was the leading country in publishing articles on this topic during the study period, contributing 28.1 % of the original articles. The US contribution percentage is similar to that observed in nuclear medicine, in which the US accounted for 26.1 % between 2002 and 2003 [5]. The US may contribute more publications because of its size
Ann Nucl Med (2014) 28:455–462
and economic resources, much as in the contributions of the country to research in other sciences. This figure is remarkably lower when comparing US productivity in other disciplines. In an analysis of the biomedical articles in the PubMed database between 1995 and 2000, Rahman et al. [23] reported that the US accounted for 40.8 % of the total articles. Comparing the output of each country for 2002 and 2012, the relative share of publications contributed by the US declined over 10 years. This phenomenon of slowing rates of US publication has been observed in the overall scientific literature [24], and there are several possible explanations for these trends, including less research funding for [25] and a greater clinical workload [26] of US investigators, particularly compared with prior time periods. Thirty-seven countries contributed to the PET literature during the study period, but in 2002, 65 % of all the articles originated from the US, Japan, Germany, the UK, or the Netherlands. PET research was more concentrated than the general nuclear medicine research, for which the leading five countries—the US, Japan, Germany, Italy, and the UK—collectively accounted for 57 % of the publications between 2002 and 2003 [5]. This result suggests that relatively few countries can perform PET research, with less research being conducted in developing countries. The relative contribution from the top five countries decreased from 65 % in 2002 to 56 % in 2012, reflecting the globalization trends in PET research over the past decade. We believe that our bibliometric study represents a useful tool for PET researchers in planning and organizing research. For example, this survey may help to interpret new perspectives of the field and may provide quantitative information about the ranking of the most successful topics. Our study also provides a useful guide for PET researchers to determine where to submit their work, because PET research is often published in other specialty journals reflecting its multidisciplinary nature. In the interpretation of the study results, several limitations should be considered. First, the articles included in the analysis were those in which ‘‘PET’’ appeared in the title, and some articles pertaining to PET in which the term was mentioned in the text of the abstract were not included in this study. However, the goal of this study was not to retrieve all the PET articles but to examine a large representative sample. Exclusion of a small number of the articles is unlikely to affect the overall results. Second, because our search was restricted to SCI/SCIE-indexed journals with an IF, published articles in journals not indexed in this database were missed. In addition, we did not include publications in languages other than English. This exclusion is based on the widespread use of English as
461
the language of science and on a practical concern for obtaining more accurate information from the articles. Third, the search was limited to three 1-year periods and the generalizability of the findings could be questioned. Given the similar changes noted in other fields of medicine, we hypothesize that the findings are valid for the identification of a trend in PET publications. Last, the assignment of the specialty and the country of origin based on the stated affiliation of the first author could potentially be problematic because some publications are the product of a collaborative effort among individuals from various departments or countries. The first author makes the greatest contribution and is most deserving of credit, and first authorship is arguably the most suitable quantitative measure of research output [27, 28]. In conclusion, the number of publications related to PET has increased rapidly over the period 2002–2012. We suggest that this study presents the characteristics and trends of PET research and provides a useful guide for PET researchers in planning and organizing their research.
References 1. Fletcher JW, Djulbegovic B, Soares HP, Siegel BA, Lowe VJ, Lyman GH, et al. Recommendations on the use of 18F-FDG PET in oncology. J Nucl Med. 2008;49:480–508. 2. Beyer T, Townsend DW, Brun T, Kinahan PE, Charron M, Roddy R, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med. 2000;41:1369–79. 3. Histed SN, Lindenberg ML, Mena E, Turkbey B, Choyke PL, Kurdziel KA. Review of functional/anatomical imaging in oncology. Nucl Med Commun. 2012;33:349–61. 4. Luukkonen T. Bibliometrics and evaluation of research performance. Ann Med. 1990;22:145–50. 5. Signore A, Annovazzi A. Scientific production and impact of nuclear medicine in Europe: how do we publish? Eur J Nucl Med Mol Imaging. 2004;31:882–6. 6. Rahman M, Sakamoto J, Fukui T. Japan’s contribution to nuclear medical research. Ann Nucl Med. 2002;16:383–5. 7. Pajares Vinardell M, Freire Macias JM. Twenty-five years of the Spanish journal of Nuclear Medicine. Bibliometric Study. Rev Esp Med Nucl. 2007;26:345–53. 8. Garfield E. The history and meaning of the journal impact factor. JAMA. 2006;295:90–3. 9. Hricak H, Choi BI, Scott AM, Sugimura K, Muellner A, von Schulthess GK, et al. Global trends in hybrid imaging. Radiology. 2010;257:498–506. 10. Greenbelt M (2011) PET imaging market summary report IMV 2011 11. Chen RC, Chu D, Chiang CH, Chou CT. Bibliometric analysis of ultrasound research trends over the period of 1991 to 2006. J Clin Ultrasound. 2009;37:319–23. 12. Miguel-Dasit A, Marti-Bonmati L, Sanfeliu P. Bibliometric analysis of the Spanish MR radiological production (2001–2007). Eur J Radiol. 2008;67:384–91. 13. Ingram TG. A cross-sectional analysis of family medicine publications in the indexed medical literature. Fam Med. 1992;24:303–6.
123
462 14. Garg AX, Iansavichus AV, Kastner M, Walters LA, Wilczynski N, McKibbon KA, et al. Lost in publication: half of all renal practice evidence is published in non-renal journals. Kidney Int. 2006;70:1995–2005. 15. Donovan AJ, Tompkins RK. Surgical research publication in a selection of research and surgical specialty journals. Surgery. 2010;147:5–12. 16. Yun EJ, Yoon DY, Kim BY, Kim YJ, Baek S, Lim KJ, et al. Where do radiologists publish their work? A comparative analysis of publications by radiologists in nonradiology journals in 2000 and 2010. AJR Am J Roentgenol. 2013;200:W560–5. 17. Seglen PO. Why the impact factor of journals should not be used for evaluating research. BMJ. 1997;314:498–502. 18. Falagas ME, Kouranos VD, Arencibia-Jorge R, Karageorgopoulos DE. Comparison of SCImago journal rank indicator with journal impact factor. FASEB J. 2008;22:2623–8. 19. Andersen J, Belmont J, Cho CT. Journal impact factor in the era of expanding literature. J Microbiol Immunol Infect. 2006;39:436–43. 20. Falagas ME, Zouglakis GM, Papastamataki PA. Trends in the impact factor of scientific journals. Mayo Clin Proc. 2006;81:1401–2. 21. Lim KJ, Yoon DY, Yun EJ, Seo YL, Baek S, Gu DH, et al. Characteristics and trends of radiology research: a survey of
123
Ann Nucl Med (2014) 28:455–462
22. 23. 24. 25. 26. 27. 28.
original articles published in AJR and Radiology between 2001 and 2010. Radiology. 2012;264:796–802. Itagaki MW, Pile-Spellman J. Factors associated with academic radiology research productivity. Radiology. 2005;237:774–80. Rahman M, Fukui T. Biomedical publication–global profile and trend. Public Health. 2003;117:274–80. Rahman M, Fukui T. A decline in the US share of research articles. N Engl J Med. 2002;347:1211–2. Fang FC, Casadevall A. NIH peer review reform–change we need, or lipstick on a pig? Infect Immun. 2009;77:929–32. Friedenberg RM. Academic medicine: boom to bust. Radiology. 2001;220:296–8. Riesenberg D, Lundberg GD. The order of authorship: who’s on first? JAMA. 1990;264:1857. Hwang SS, Song HH, Baik JH, Jung SL, Park SH, Choi KH, et al. Researcher contributions and fulfillment of ICMJE authorship criteria: analysis of author contribution lists in research articles with multiple authors published in radiology. International Committee of Medical Journal Editors. Radiology. 2003;226: 16–23.
ORIGINAL ARTICLE
Characteristics and trends of research on positron emission tomography: a bibliometric analysis, 2002–2012 Sora Baek • Dae Young Yoon • Kyung Joon Min Kyoung Ja Lim • Young Lan Seo • Eun Joo Yun
•
Received: 24 January 2014 / Accepted: 27 February 2014 / Published online: 12 March 2014 Ó The Japanese Society of Nuclear Medicine 2014
Abstract Objective We performed a bibliometric analysis of scientific publications focused on positron emission tomography (PET) over a 10-year period. Methods The MEDLINE and ISI Web of Knowledge databases were searched for English language original articles focused on PET in SCI/SCIE-indexed journals in 2002, 2007, and 2012. We selected the documents with titles that included ‘‘PET’’ or ‘‘positron emission.’’ The following information was obtained from each article: journal (year of publication, title, subject category, and impact factor), subspecialty, imaging modality used, tracer, species, sample size, number of authors, affiliation of the first author, declared funding, and country of origin. Results The yearly publication on PET increased from 547 (2002) to 986 (2007), and 1838 (2012). A total of 1753 (52.0 %) articles were published in journals in the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ category, 1512 (44.9 %) were in the subspecialty of oncology, 3245 (96.3 %) used PET or PET/CT, 1698 (50.4 %) used 2-fluoro-2-deoxy-D-glucose (FDG) as the radiotracer, 2378 (70.5 %) were human studies, 1294 (38.4 %) had a sample size of \20, 1674 (49.7 %) had [7 authors, 779 (23.1 %) were written by a first author from a department of nuclear medicine, and 1337 (39.7 %) were supported by
S. Baek Department of Nuclear Medicine, Kangdong Seong-Sim Hospital, Hallym University College of Medicine, 445 Gil-dong Kangdong-Gu, Seoul 134-701, Korea D. Y. Yoon (&) K. J. Min K. J. Lim Y. L. Seo E. J. Yun Department of Radiology, Kangdong Seong-Sim Hospital, Hallym University College of Medicine, 445 Gil-dong Kangdong-Gu, Seoul 134-701, Korea e-mail: [email protected]
government funding. The United States published 948 studies (28.1 %) followed by Japan (345, 10.2 %) and Germany (335, 9.9 %). In the time trend analysis oncology subspecialty, PET/MR as the imaging modality, FDG as the tracer, sample size [50, number of authors [7, radiology department affiliation of the first author, and government funding exhibited significantly positive trends. Conclusions The number of publication concerning PET has increased rapidly over the last decade. This bibliometric analysis revealed characteristics and trends of current PET research that provides useful information to researchers. Keywords
PET Research Publications Bibliometrics
Introduction Positron emission tomography (PET) is a diagnostic modality in nuclear medicine that enables a three-dimensional imaging of functional changes in the body [1]. Recently, PET has emerged as one of the most promising imaging modalities in clinical medicine. PET has been used in combination with high-resolution anatomical imaging modalities (i.e., CT or MR) to provide functional and anatomical information simultaneously [2, 3]. Many investigators, particularly in the last decade, have published articles on PET, reflecting the increased clinical use, technological progress, and expanded clinical applications of this technique. Bibliometry is a research method used to evaluate the quantity and quality of scientific research production within a specific field or a specific topic over a given period [4]. Bibliometric analysis provides insight into research patterns and trends in scientific publications. Although
123
456
several bibliometric studies have been conducted in the field of nuclear medicine [5–7], to the best of our knowledge, no study has explored the characteristics of research in PET. The purpose of this study was as follows: (1) to quantify the worldwide output, (2) to assess the characteristics, and (3) to determine the trends of PET research over a 10-year period (2002–2012).
Methods Our study was a retrospective bibliometric analysis that did not involve human subjects and was exempt from Institutional Review Board approval. Study search and data collection The first step consisted of identifying all the articles focused on PET during three 1-year periods (2002, 2007, and 2012). The National Library of Medicine’s PubMed database (URL: http://www.ncbi.nlm.nih.gov/pubmed/) was searched between February and April 2013 to identify the primary studies of interest. We selected documents including the words ‘‘PET’’ or ‘‘positron emission’’ in the title. The term ‘‘positron emission’’ was selected to widen our search to include articles with a variety of words in the title (‘‘positron emission tomography’’, ‘‘positron emission tomographic’’, ‘‘positron emission mammography’’). To establish the characteristics of the articles focused on PET, only original articles were included in the analysis. Original articles were considered to be documents that investigated clearly stated objectives or hypotheses and contained specifically articulated methods and results sections. Other document types such as case reports, review articles, commentaries, editorials, letters, technical notes, quizzes, lectures, book reviews, and news were excluded from the analysis. We included only articles written in English; articles published in other languages were excluded. Finally, we excluded articles with a focus on pet (a domesticated animal kept for companionship or amusement). Abstracting data on journals We accessed the Institute for Scientific Information (ISI) Web of Knowledge (URL: http://www.isiwebofknowledge. com), a database of more than 8400 journals in more than 170 scientific disciplines, for a list of all the journals indexed by the Science Citation Index (SCI) and the Science Citation Index Expanded (SCIE). The articles published in journals that were not indexed by SCI/SCIE were excluded from the analysis.
123
Ann Nucl Med (2014) 28:455–462
The analyzed journals (SCI/SCIE-indexed journals that published at least one original article focused on PET) were assigned to one of a number of subject categories (169, 171, and 176 categories in 2002, 2007, and 2012, respectively) based on information in the ISI Web of Knowledge-Journal Citation Reports (JCR). According to the ISI Web of Knowledge-JCR, a number of journals were listed under more than one subject category, depending on the content. Journals that were in more than one subject category were manually reviewed and assigned to one category based on the title of the journal, information contained in ‘‘Instructions to Authors,’’ and the table of contents of a sample issue. Decisions about assigning these journals were reached by consensus between two study investigators (S.B., a board-certified nuclear medicine physician with 3 years of experience and D.Y.Y., a board-certified radiologist with 18 years of experience). The journal impact factor (IF) of each journal was obtained for the year of publication (2002, 2007, and 2012) from the ISI Web of Knowledge-JCR. The IF of a journal for a given year is calculated by dividing the number of citations in the JCR year by the total number of articles published in the past two years [8]. Abstracting data on articles The following information was abstracted from each article for analysis: (1) the subspecialty [oncology, cardiovascular disease, inflammation/infection, metabolism, neurology, physics/instrumental, radiopharmaceutical (tracer), molecular imaging, or others (not conforming to one of the above categories)]; (2) the imaging modality used (PET or PET/ CT, PET/MR, or others); (3) the tracer [FDG (2-fluoro-2deoxy-D-glucose), others (including multiple tracers), or none]; (4) the species (human, animal, or others); (5) the sample size (none, \20, 21–50, or [50); (6) the number of authors (\4, 4–7, or [7); (7) the affiliation of the first author [radiology (including radiology, imaging center, and other imaging-related specialties), nuclear medicine, medicine or related specialties, surgery or related specialties, or other]; (8) the declared funding (government funding, other resources, or none); and (9) the country of origin. For the articles arising from a collaboration involving more than one country, the country of the first author was considered the country of origin of the paper. If the first author had an affiliation with more than one country, the country of origin of the paper was determined by the country of the corresponding author. Two investigators (S.B. and D.Y.Y.) initially reviewed the same 200 articles independently to ensure consistency of the data abstraction. Any disagreements were resolved in a consensus meeting; however, only two (1.0 %)
Ann Nucl Med (2014) 28:455–462
disagreements occurred with regard to the subspecialty field of research. After initial pilot abstraction, the total number of articles was divided randomly into 2 samples and manually reviewed by the investigators. The questionable cases were decided by consensus of the 2 study investigators.
457 Table 1 Comparison of SCI/SCIE-indexed journals which published original articles focused on PET in 2002, 2007, and 2012 2002
2007
2012
No. of articles
547
986
1838
No. of journals
178
247
378 3.8 ± 2.6*
Journal IF
Statistical analysis The data were recorded by publication year (2002, 2007, and 2012) and downloaded into a spreadsheet for descriptive statistics using Microsoft Excel version 11.0 (Microsoft Corporation, Redmond, WA, USA). The journal IFs were compared using Student’s t test and the other categorical variables were compared using the Chi-square test. The statistical analyses were performed using SPSS software (SPSS version 19.0 for Windows; SPSS, Chicago, IL, USA) and a p value of \0.05 was considered to be statistically significant.
Results A total of 547, 986, and 1838 original articles focused on PET were written in English and published in SCI/SCIEindexed journals in 2002, 2007, and 2012, respectively. The absolute number of articles increased 3.4-fold from 2002 through 2012 (Table 1). The articles were published in 515 scientific journals. Table 2 presents a list of the 20 journals with the highest number of articles focused on PET. Regarding the JCR subject categories of the journals, 1753 articles (52.0 %) were published in journals under the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ subject category. Other subject categories included ‘‘Oncology’’ (n = 269, 8.0 %) and ‘‘Neurosciences’’ (n = 254, 7.5 %) (Table 3). The mean IF of the journals that published PET research increased from 3.3 in 2002 to 3.8 in 2007, a statistically significant increase (p \ 0.05). No significant change in the IF was observed from 2007 to 2012 (Table 1). ‘Oncology’ was found to be the most productive field (n = 1512, 44.9 %), followed by ‘neurology’ (n = 625, 18.5 %), ‘physics/instrumental’ (n = 325, 9.6 %), and ‘radiopharmaceutical’ (n = 237, 7.0 %). In the time trend analysis, the ‘oncology’, ‘radiopharmaceutical’ and ‘molecular imaging’ subspecialties demonstrated a significant increase, whereas the ‘neurology’ subspecialty demonstrated a significant decrease (Table 4). Other variables including the imaging modality used, tracer, species, sample size, number of authors, affiliation of the first author, and declared funding are summarized in Table 4. The variables that showed a statistically significant increase over time included the following: PET/MR in
Mean ± SD
3.3 ± 2.2
3.8 ± 2.4*
25 percentile Journal IF
1.924
2.209
2.247
50 percentile (median) journal IF
2.930
3.561
3.095
75 percentile journal IF
4.587
5.457
5.114
IF impact factor * Statistically significant difference compared to 2002 (p = 0.001, Student’s t test)
imaging modality, FDG in tracer, sample size [ 50, number of authors [ 7, department of radiology in the affiliation of the first author, and government funding. The variables that showed a statistically significant decrease over time included the following: others in tracer, sample size \ 20, number of authors \ 4 and 4–7, nuclear medicine in the affiliation of the first author, and other resources of declared funding (Table 4). Table 5 shows the data from the 20 most productive countries in PET research. The United States (US) was dominant with the highest number of publications (n = 948, 28.1 %) followed by Japan (n = 345, 10.2 %), Germany (n = 335, 9.9 %), South Korea (n = 178, 5.3 %), and the United Kingdom (UK) (n = 174, 5.2 %). The articles from these five countries accounted for 58.7 % of the PET research. South Korea, Italy, China, Taiwan, Turkey, and India showed a statistically significant increase over the analyzed period in the proportion of articles, whereas US, Germany, the UK, and Finland exhibited a significant decline in the percentage of articles (Table 5).
Discussion PET is an imaging technique that provides detailed information regarding the molecular and metabolic changes of an organ. In the mid-1980s, with the development of commercial PET scanners, PET moved from the research field to clinical practice. A major advance in PET scanning occurred in the early 1990s with the introduction of the PET/CT scanner. The hybrid of PET and CT imaging, simultaneously showing metabolic and structural images, enhances the ability of a physician to localize abnormalities accurately. The addition of integrated dual-modality imaging and the development of new radiotracers have accelerated the rapid growth of PET during recent years [9]. It is estimated that in the US, the number of PET/CT
123
458 Table 2 Top 10 SCI/SCIEindexed journals with the highest number of articles focused on PET in 2002, 2007, and 2012
Table 3 Top 10 subject categories with the highest number of articles focused on PET in 2002, 2007, and 2012
Values in parentheses are percentages Ranking based on the total number of original articles published in 2002, 2007, and 2012 :
Statistically significant increase between 2002 and 2012 (Chi-square test)
;
Statistically significant decrease between 2002 and 2012 (Chi-square test)
Ann Nucl Med (2014) 28:455–462
Rank
2002
2007
Journal
No. of articles
Journal
No. of articles
Journal
No. of articles
1
Eur J Nucl Med Mol Imaging
65
J Nucl Med
110
J Nucl Med
153
2
J Nucl Med
49
Eur J Nucl Med Mol Imaging
106
Eur J Nucl Med Mol Imaging
135
3
Nucl Med Biol
28
Phys Med Biol
30
4
Neuroimage
25
Neuroimage
29
Ann Nucl Med
55
5
Phys Med Biol
17
Nucl Med Commun
28
Clin Nucl Med
52
6
Synapse
16
Int J Radiat Oncol Biol Phys
27
Phys Med Biol
51
7
Acta Radiol
14
Nucl Med Biol
26
Eur J Radiol
49
8
J Cereb Blood Flow Metab
13
Mol Imaging Biol
22
Nucl Med Biol
49
9
13
Ann Nucl Med
21
9
J Cereb Blood Flow Metab
18
Int J Radiat Oncol Biol Phys Mol Imaging Biol
47
10
Nucl Med Commun Ann Nucl Med
Rank
Category
Nucl Med Commun
56
42
No. of Articles 2002
2007
2012
Total
239 (43.7)
518 (52.5)
996 (54.2)
1753 (52.0)
1
Radiology, Nuclear Medicine & Medical Imaging:
2
Oncology
45 (8.2)
90 (9.1)
134 (7.3)
269 (8.0)
3
Neurosciences;
77 (14.1)
87 (8.8)
90 (4.9)
254 (7.5)
4
Neuroimaging;
27 (4.9)
33 (3.3)
48 (2.6)
108 (3.2)
5
Clinical Neurology;
27 (4.9)
29 (2.9)
42 (2.3)
98 (2.9)
5
Chemistry, Inorganic & Nuclear
8 (1.5)
25 (2.5)
52 (2.9)
85 (2.5)
6 7
Pharmacology & Pharmacy Surgery
12 (2.2) 14 (2.6)
20 (2.0) 12 (1.2)
42 (2.3) 36 (2.0)
74 (2.2) 62 (1.8)
8
Hematology
3 (0.5)
19 (1.9)
31 (1.7)
53 (1.6)
9
Otorhinolaryngology
5 (0.9)
13 (1.3)
33 (1.8)
51 (1.5)
10
Cardiac & Cardiovascular Systems
10 (1.8)
19 (1.9)
21 (1.1)
50 (1.5)
systems increased dramatically by more than 10-fold (from approximately 200 to 2085) and the number of PET examinations increased nearly sevenfold (from approximately 250000 to approximately 1.74 million) between 2001 and 2010 [10]. In the evidence-based model of clinical practice, research precedes clinical adoption. One of the striking findings of our study was the rapid growth in the number of publications on PET research during the last decade. Overall, the total number of publications increased 3.4-fold from 2002 through 2012. This marked growth in the scientific publications in this field suggests that PET is an expanding and promising field of research. The increase in
123
2012
PET publications appears to be more pronounced than those observed in other imaging modalities (i.e., ultrasonography or MR) [11, 12]. This growth may reflect the following: (1) a rapidly increasing worldwide commercial manufacture of PET/CT systems, (2) a widening application of PET resulting from technological development and clinical introduction of new pharmacological agents, and (3) an increasing knowledge and scientific interest in PET. Our study showed that there are a variety of journals that publish PET research. The articles on PET were predominantly published in imaging journals (journals in the ‘‘Radiology, Nuclear Medicine & Medical Imaging’’ category). An analysis of the journals with the highest number
Ann Nucl Med (2014) 28:455–462
459
Table 4 Characteristics and trends of original articles focused on PET in 2002, 2007, and 2012
Number of Articles 2002
2007
2012
Total
Subspecialty Oncology:
191 (34.9)
436 (44.2)
885 (48.2)
1512 (44.9)
Cardiovascular disease
28 (5.1)
56 (5.7)
129 (7.0)
213 (6.3)
Inflammation/infection
13 (2.4)
35 (3.5)
58 (3.2)
106 (3.1)
Metabolism
19 (3.5)
37 (3.8)
42 (2.3)
98 (2.9)
Neurology;
193 (35.3)
216 (21.9)
216 (11.8)
625 (18.5)
Physics/Instrumental
52 (9.5)
96 (9.7)
177 (9.6)
325 (9.6)
Radiopharmaceutical (tracer):
32 (5.9)
41 (4.2)
164 (8.9)
237 (7.0)
Molecular imaging:
13 (2.4)
59 (6.0)
77 (4.2)
149 (4.4)
6 (1.1)
10 (1.0)
90 (4.9)
106 (3.1)
534 (97.6)
951 (96.5)
1760 (95.8)
3245 (96.3)
2 (0.4)
10 (1.0)
52 (2.8)
64 (1.9)
11 (2.0)
25 (2.5)
26 (1.4)
62 (1.8)
236 (43.1)
493 (50.0)
969 (52.7)
1698 (50.4)
278 (50.8)
432 (43.8)
761 (41.4)
1471 (43.6)
33 (6.0)
61 (6.2)
108 (5.9)
202 (6.0)
Human
390 (71.3)
696 (70.6)
1292 (70.3)
2378 (70.5)
Animal
80 (14.6)
159 (16.1)
255 (13.9)
494 (14.7)
Others
77 (14.1)
131 (13.3)
291 (15.8)
496 (14.7)
Others: Imaging modality used PET or PET/CT PET/MR: Others Tracer FDG*, Others
: ;
None Species
Sample size None
64 (11.7)
132 (13.4)
232 (12.6)
428 (12.7)
\20;
296 (54.1)
391 (39.7)
607 (33.0)
1294 (38.4)
21–50
131 (23.9)
244 (24.7)
456 (24.8)
831 (24.7)
:
56 (10.2)
219 (22.2)
543 (29.5)
818 (24.3)
\4;
74 (13.5)
66 (6.7)
118 (6.4)
258 (7.7)
4–7;
260 (47.5)
441 (44.7)
738 (40.2)
1439 (42.7)
[7:
213 (38.9)
479 (48.6)
982 (53.4)
1674 (49.7)
[50 Number of authors
Affiliation of the first author Radiology (imaging center):
Values in parentheses are percentages * 2-fluoro-2-deoxy-D-glucose :
Statistically significant increase between 2002 and 2012 (Chi-square test)
;
Statistically significant decrease between 2002 and 2012 (Chi-square test)
89 (16.3)
242 (24.5)
393 (21.4)
724 (21.5)
Nuclear medicine;
155 (28.3)
255 (25.9)
369 (20.1)
779 (23.1)
Medicine or related specialties
126 (23.0)
223 (22.6)
374 (20.3)
723 (21.4)
56 (10.2)
82 (8.3)
173 (9.4)
311 (9.2)
121 (22.1)
184 (18.7)
529 (28.8)
834 (24.7)
Government funding:
196 (35.8)
379 (38.4)
762 (41.5)
1337 (39.7)
Other resources;
109 (19.9)
158 (16.0)
240 (13.1)
507 (15.0)
None
242 (44.2)
449 (45.5)
836 (45.5)
1527 (45.3)
Surgery or related specialties Others
:
Declared funding
Total
of articles in this field showed a prevalence of articles in journals focused on ‘‘nuclear medicine’’. Five of the 10 journals with the highest number of articles contained the word ‘‘nuclear’’ in the journal title in 2002, 2007, and 2012 (Table 2). Approximately one-half (48 %) of the articles
547
986
1838
3371
on PET were published in non-imaging journals (journals under the category of other than ‘‘Radiology, Nuclear Medicine & Medical Imaging’’). This tendency primarily reflects the impact of PET on basic and clinical specialties. Fifty-five percent of the articles on PET were primarily
123
460 Table 5 Top 20 countries ranked according to total number of original articles focused on PET in 2002, 2007, and 2012
Ann Nucl Med (2014) 28:455–462
Rank
Number of Articles 2002
2007
2012
Total 948 (28.1)
1
United States;
163 (29.8)
319 (32.4)
466 (25.4)
2
Japan
62 (11.3)
98 (9.9)
185 (10.1)
345 (10.2)
3
Germany;
61 (11.2)
120 (12.2)
154 (8.4)
335 (9.9)
4
South Korea:
11 (2.0)
42 (4.3)
125 (6.8)
178 (5.3)
a, ;
5
United Kingdom
45 (8.2)
35 (3.5)
94 (5.1)
174 (5.2)
6
France
18 (3.3)
52 (5.3)
81 (4.4)
151 (4.5)
Values in parentheses are percentages
7
Italy:
10 (1.8)
30 (3.0)
97 (5.3)
137 (4.1)
8
The Netherlands
23 (4.2)
38 (3.9)
73 (4.0)
134 (4.0)
Ranking based on the total number of original articles published in 2002, 2007, and 2012
9
Canada
18 (3.3)
34 (3.4)
55 (3.0)
107 (3.2)
10
Chinab,
1 (0.2)
20 (2.0)
83 (4.5)
104 (3.1)
11 12
Switzerland Sweden
22 (4.0) 21 (3.8)
28 (2.8) 22 (2.2)
44 (2.4) 43 (2.3)
94 (2.8) 86 (2.6)
13
Denmark
12 (2.2)
20 (2.0)
42 (2.3)
74 (2.2)
14
Belgium
13 (2.4)
18 (1.8)
31 (1.7)
62 (1.8)
15
Australia
12 (2.2)
10 (1.0)
33 (1.8)
55 (1.6)
5 (0.9)
10 (1.0)
38 (2.1)
53 (1.6)
3 (0.5)
11 (1.1)
31 (1.7)
45 (1.3)
15 (1.5)
13 (0.7)
41 (1.2)
a
Includes articles originating from England, Scotland, Wales, and Northern Ireland
b
Includes articles originating from Hong Kong
: Statistically significant increase between 2002 and 2012 (Chi-square test) ; Statistically significant decrease between 2002 and 2012 (Chi-square test)
:
:
16
Taiwan
17
Spain
18
Finland;
13 (2.4)
19
Turkey:
1 (0.2)
9 (0.9)
27 (1.5)
37 (1.1)
20
India:
0 (0)
1 (0.1)
34 (1.8)
35 (1.0)
authored by non-radiologist/non-nuclear medicine physicians. Publication across disciplines is a current trend in scientific literature [13–16]. The most commonly used indicator to evaluate the relative quality of scientific journals is the IF of a journal, but it is not, however, a perfect tool to measure the quality of journals [17, 18]. The IF of biomedical journals has generally shown a tendency to rise, according to the expansion in the size of the relevant literature [19, 20]. In our study, the IF of journals that published PET research did not change significantly from 2007 to 2012. Considering the overall increase in the IF of the journal, this finding indicates that the scientific impact of PET research decreased between 2007 and 2012. The proportion of PET articles published in the nuclear medicine journal with the highest impact factor (Journal of Nuclear Medicine) decreased from 11.2 % in 2007 to 8.3 % in 2012. The analysis of the distribution of papers according to the subspecialty field revealed that oncology accounted for almost one-half of PET research. This finding is predominantly based on the emerging role of PET or PET/CT using FDG in the detection of primary and recurrent malignant tumors. A total of 45 % of the articles on PET were primarily authored by radiologists or nuclear medicine physicians between 2002 and 2012. The percentage of articles written by radiologists (or physicians of the imaging center)
123
significantly increased, whereas the percentage of articles written by nuclear medicine physicians significantly decreased. The reason for this disparity is unclear and likely multifactorial. First, department of nuclear medicine has been partially or fully separated from radiology departments in most countries. In recent years, the borders between the two specialties have not been clearly defined (e.g., the emergence of imaging centers). Second, the use of hybrid machines such as PET/CT or PET/MR has resulted in a merger of nuclear medicine and radiology in PET research. The overall funding rate of 54.7 % in PET research over the entire study period is substantially higher than that reported for general radiology research (26.9 %) [21]. The positive association between funding and research productivity is well established [22]. We found that government funding for PET research increased nearly 6 % during the study period whereas the overall proportion of funding did not change significantly. Increased government funding such as National Institutes of Health may have led to an increase in financial support for higher-level research. The US was the leading country in publishing articles on this topic during the study period, contributing 28.1 % of the original articles. The US contribution percentage is similar to that observed in nuclear medicine, in which the US accounted for 26.1 % between 2002 and 2003 [5]. The US may contribute more publications because of its size
Ann Nucl Med (2014) 28:455–462
and economic resources, much as in the contributions of the country to research in other sciences. This figure is remarkably lower when comparing US productivity in other disciplines. In an analysis of the biomedical articles in the PubMed database between 1995 and 2000, Rahman et al. [23] reported that the US accounted for 40.8 % of the total articles. Comparing the output of each country for 2002 and 2012, the relative share of publications contributed by the US declined over 10 years. This phenomenon of slowing rates of US publication has been observed in the overall scientific literature [24], and there are several possible explanations for these trends, including less research funding for [25] and a greater clinical workload [26] of US investigators, particularly compared with prior time periods. Thirty-seven countries contributed to the PET literature during the study period, but in 2002, 65 % of all the articles originated from the US, Japan, Germany, the UK, or the Netherlands. PET research was more concentrated than the general nuclear medicine research, for which the leading five countries—the US, Japan, Germany, Italy, and the UK—collectively accounted for 57 % of the publications between 2002 and 2003 [5]. This result suggests that relatively few countries can perform PET research, with less research being conducted in developing countries. The relative contribution from the top five countries decreased from 65 % in 2002 to 56 % in 2012, reflecting the globalization trends in PET research over the past decade. We believe that our bibliometric study represents a useful tool for PET researchers in planning and organizing research. For example, this survey may help to interpret new perspectives of the field and may provide quantitative information about the ranking of the most successful topics. Our study also provides a useful guide for PET researchers to determine where to submit their work, because PET research is often published in other specialty journals reflecting its multidisciplinary nature. In the interpretation of the study results, several limitations should be considered. First, the articles included in the analysis were those in which ‘‘PET’’ appeared in the title, and some articles pertaining to PET in which the term was mentioned in the text of the abstract were not included in this study. However, the goal of this study was not to retrieve all the PET articles but to examine a large representative sample. Exclusion of a small number of the articles is unlikely to affect the overall results. Second, because our search was restricted to SCI/SCIE-indexed journals with an IF, published articles in journals not indexed in this database were missed. In addition, we did not include publications in languages other than English. This exclusion is based on the widespread use of English as
461
the language of science and on a practical concern for obtaining more accurate information from the articles. Third, the search was limited to three 1-year periods and the generalizability of the findings could be questioned. Given the similar changes noted in other fields of medicine, we hypothesize that the findings are valid for the identification of a trend in PET publications. Last, the assignment of the specialty and the country of origin based on the stated affiliation of the first author could potentially be problematic because some publications are the product of a collaborative effort among individuals from various departments or countries. The first author makes the greatest contribution and is most deserving of credit, and first authorship is arguably the most suitable quantitative measure of research output [27, 28]. In conclusion, the number of publications related to PET has increased rapidly over the period 2002–2012. We suggest that this study presents the characteristics and trends of PET research and provides a useful guide for PET researchers in planning and organizing their research.
References 1. Fletcher JW, Djulbegovic B, Soares HP, Siegel BA, Lowe VJ, Lyman GH, et al. Recommendations on the use of 18F-FDG PET in oncology. J Nucl Med. 2008;49:480–508. 2. Beyer T, Townsend DW, Brun T, Kinahan PE, Charron M, Roddy R, et al. A combined PET/CT scanner for clinical oncology. J Nucl Med. 2000;41:1369–79. 3. Histed SN, Lindenberg ML, Mena E, Turkbey B, Choyke PL, Kurdziel KA. Review of functional/anatomical imaging in oncology. Nucl Med Commun. 2012;33:349–61. 4. Luukkonen T. Bibliometrics and evaluation of research performance. Ann Med. 1990;22:145–50. 5. Signore A, Annovazzi A. Scientific production and impact of nuclear medicine in Europe: how do we publish? Eur J Nucl Med Mol Imaging. 2004;31:882–6. 6. Rahman M, Sakamoto J, Fukui T. Japan’s contribution to nuclear medical research. Ann Nucl Med. 2002;16:383–5. 7. Pajares Vinardell M, Freire Macias JM. Twenty-five years of the Spanish journal of Nuclear Medicine. Bibliometric Study. Rev Esp Med Nucl. 2007;26:345–53. 8. Garfield E. The history and meaning of the journal impact factor. JAMA. 2006;295:90–3. 9. Hricak H, Choi BI, Scott AM, Sugimura K, Muellner A, von Schulthess GK, et al. Global trends in hybrid imaging. Radiology. 2010;257:498–506. 10. Greenbelt M (2011) PET imaging market summary report IMV 2011 11. Chen RC, Chu D, Chiang CH, Chou CT. Bibliometric analysis of ultrasound research trends over the period of 1991 to 2006. J Clin Ultrasound. 2009;37:319–23. 12. Miguel-Dasit A, Marti-Bonmati L, Sanfeliu P. Bibliometric analysis of the Spanish MR radiological production (2001–2007). Eur J Radiol. 2008;67:384–91. 13. Ingram TG. A cross-sectional analysis of family medicine publications in the indexed medical literature. Fam Med. 1992;24:303–6.
123
462 14. Garg AX, Iansavichus AV, Kastner M, Walters LA, Wilczynski N, McKibbon KA, et al. Lost in publication: half of all renal practice evidence is published in non-renal journals. Kidney Int. 2006;70:1995–2005. 15. Donovan AJ, Tompkins RK. Surgical research publication in a selection of research and surgical specialty journals. Surgery. 2010;147:5–12. 16. Yun EJ, Yoon DY, Kim BY, Kim YJ, Baek S, Lim KJ, et al. Where do radiologists publish their work? A comparative analysis of publications by radiologists in nonradiology journals in 2000 and 2010. AJR Am J Roentgenol. 2013;200:W560–5. 17. Seglen PO. Why the impact factor of journals should not be used for evaluating research. BMJ. 1997;314:498–502. 18. Falagas ME, Kouranos VD, Arencibia-Jorge R, Karageorgopoulos DE. Comparison of SCImago journal rank indicator with journal impact factor. FASEB J. 2008;22:2623–8. 19. Andersen J, Belmont J, Cho CT. Journal impact factor in the era of expanding literature. J Microbiol Immunol Infect. 2006;39:436–43. 20. Falagas ME, Zouglakis GM, Papastamataki PA. Trends in the impact factor of scientific journals. Mayo Clin Proc. 2006;81:1401–2. 21. Lim KJ, Yoon DY, Yun EJ, Seo YL, Baek S, Gu DH, et al. Characteristics and trends of radiology research: a survey of
123
Ann Nucl Med (2014) 28:455–462
22. 23. 24. 25. 26. 27. 28.
original articles published in AJR and Radiology between 2001 and 2010. Radiology. 2012;264:796–802. Itagaki MW, Pile-Spellman J. Factors associated with academic radiology research productivity. Radiology. 2005;237:774–80. Rahman M, Fukui T. Biomedical publication–global profile and trend. Public Health. 2003;117:274–80. Rahman M, Fukui T. A decline in the US share of research articles. N Engl J Med. 2002;347:1211–2. Fang FC, Casadevall A. NIH peer review reform–change we need, or lipstick on a pig? Infect Immun. 2009;77:929–32. Friedenberg RM. Academic medicine: boom to bust. Radiology. 2001;220:296–8. Riesenberg D, Lundberg GD. The order of authorship: who’s on first? JAMA. 1990;264:1857. Hwang SS, Song HH, Baik JH, Jung SL, Park SH, Choi KH, et al. Researcher contributions and fulfillment of ICMJE authorship criteria: analysis of author contribution lists in research articles with multiple authors published in radiology. International Committee of Medical Journal Editors. Radiology. 2003;226: 16–23.
